In this paper we propose a feature-based analysis of Cappadocian Greek nominal inflection. In particular, based on the formal description of Standard Greek nominal inflection by Ralli (2000), we identify the inflectional formatives and analyse them as bundles of features expressing number, case, animacy and inflectional class. The feature specification of each ending determines its distribution in the relevant variety. Similarly, inflectional classes are defined on the basis of (a) the compatibility of a given noun with an inflectional pattern and (b) the base alternations it exhibits. Such an approach has the advantage of accounting for the variation in base formation and for the agglutinative inflectional patterns by means of systematic base allomorphy and default formatives.

The optional use of morphology attested in second language learners has been attributed either to a representational deficit or to a ‘surface’ problem with respect to the realization of inflectional affixes. In this article we contribute to this issue by providing empirical data from the early interlanguage of Greek learners of Turkish. Three experiments have been conducted, a cloze task, a sentence picture matching task and an on-line grammaticality judgement task, in order to investigate case morphology and its interaction with word order constraints.The findings of all three experiments point towards a variable use of case morphology, which is also observed in previous studies of Turkish as a second language (L2). Moreover, they show clearly that the learners face difficulties with non-canonical word orders as well as with the interaction of word order constraints and Case. On the other hand, the learners performed well on verbal inflections. On the basis of these findings, we argue that the developmental patterns in the early stages of L2 acquisition cannot be attributed to a global lack of functional categories but rather to more localized difficulties, which seem to be related to (a) whether the features in the L2 are grammaticalized in the first language and (b) the way these features are encoded in the morphosyntax of the first language. Moreover, we claim that processing factors and the specific properties of the morphological paradigms affect L2 development.

Case matching effects involve situations where a nominal element appears in a case that is not predicted by its syntactic status in its clause, but rather matches the case requirements of an external element. Such constructions pose serious problems for syntactic accounts of case in terms of a syntactic Case feature, because they involve a situation of case conflict/mismatch. In this article, I address the issue of case assignment/realization by examining case attraction phenomena in Greek free relative clauses. In particular, I suggest an analysis that builds on the idea that case categories are not primitives, but rather they can be decomposed in bundles of features and I propose a division of labour between narrow syntax and Morphological Structure as far as case assignment/realization is concerned. Case assignment takes place in narrow syntax as a licensing device (abstract case), but it refers only to those features that are relevant to the distinction between structural and inherent case. The full specification of the case feature bundle takes place in the Morphological Structure as a result of the application of specific case assignment algorithms defined in terms of case domains and hierarchies. At a theoretical level, such a hypothesis has the benefit of incorporating the insights about the role of case determination at Morphological Structure, as well as maintaining the well-established notion of abstract case as a licensing device of narrow syntax.